def train(self, adj_list, fea_list, y_train, y_val, y_test, train_mask, val_mask, test_mask, y_all, all_mask, rawlabels, needtSNE=False, rawFeature=[]):
prec, rec, f1 = 0.0, 0.0, 0.0
nb_nodes = fea_list[0].shape[0]
ft_size = fea_list[0].shape[1]
nb_classes = y_train.shape[1]
# nb_classes = len(set(rawlabels))
# adj = adj.todense()
# features = features[np.newaxis] # [1, nb_node, ft_size]
fea_list = [fea[np.newaxis] for fea in fea_list]
adj_list = [adj[np.newaxis] for adj in adj_list]
y_train = y_train[np.newaxis]
y_val = y_val[np.newaxis]
y_test = y_test[np.newaxis]
y_all = y_all[np.newaxis]
train_mask = train_mask[np.newaxis]
val_mask = val_mask[np.newaxis]
test_mask = test_mask[np.newaxis]
all_mask = all_mask[np.newaxis]
biases_list = [process.adj_to_bias(adj, [nb_nodes], nhood=1) for adj in adj_list]
print('build graph...')
with tf.Graph().as_default():
with tf.name_scope('input'):
metric_ftr_in = tf.placeholder(dtype=tf.float32, shape=(nb_nodes, ft_size), name='metric_ftr_in')
ftr_in_list = [tf.placeholder(dtype=tf.float32,
shape=(batch_size, nb_nodes, ft_size),
name='ftr_in_{}'.format(i))
for i in range(len(fea_list))]
bias_in_list = [tf.placeholder(dtype=tf.float32,
shape=(batch_size, nb_nodes, nb_nodes),
name='bias_in_{}'.format(i))
for i in range(len(biases_list))]
lbl_in = tf.placeholder(dtype=tf.int32, shape=(
batch_size, nb_nodes, nb_classes), name='lbl_in')
msk_in = tf.placeholder(dtype=tf.int32, shape=(batch_size, nb_nodes),
name='msk_in')
attn_drop = tf.placeholder(dtype=tf.float32, shape=(), name='attn_drop')
ffd_drop = tf.placeholder(dtype=tf.float32, shape=(), name='ffd_drop')
is_train = tf.placeholder(dtype=tf.bool, shape=(), name='is_train')
# forward
logits, final_embedding, att_val, centers_embed, test_final_embeed = model.inference(ftr_in_list, nb_classes, nb_nodes, is_train,
attn_drop, ffd_drop,
bias_mat_list=bias_in_list,
hid_units=hid_units, n_heads=n_heads, features=fea_list, labels=rawlabels,
residual=residual, activation=nonlinearity, feature_size=ft_size)
# final_embedding: checkout Tensor("Sum:0", shape=(286, 64), dtype=float32)
# logits: checkout Tensor("ExpandDims_3:0", shape=(1, 286, 30), dtype=float32)
# cal masked_loss
# lab_list = tf.placeholder(dtype=tf.float32, shape=(nb_nodes, ), name='lab_list')
# ftr_resh = tf.placeholder(dtype=tf.float32, shape=(nb_nodes, ft_size), name='ftr_resh')
log_resh = tf.reshape(logits, [-1, nb_classes])
lab_resh = tf.reshape(lbl_in, [-1, nb_classes])
msk_resh = tf.reshape(msk_in, [-1])
print ("final_embedding: checkout", final_embedding)
print ("logits: checkout", logits)
print ("log_resh: checkout", log_resh)
# print ("ftr_resh: ", ftr_resh)
print ("lab_resh: ", lab_resh)
print ("fea_list: ", fea_list)
print ("centers_embed: ", centers_embed)
print ("batch_size, nb_nodes, nb_classes, ft_size", batch_size, nb_nodes, nb_classes, ft_size)
osm_caa_loss = OSM_CAA_Loss(batch_size=nb_nodes)
osm_loss = osm_caa_loss.forward
# final_embedding: checkout Tensor("Sum:0", shape=(286, 64), dtype=float32)
# logits: checkout Tensor("ExpandDims_3:0", shape=(1, 286, 30), dtype=float32)
# log_resh: checkout Tensor("Reshape:0", shape=(286, 30), dtype=float32)
# ftr_resh: Tensor("ftr_resh:0", shape=(286, 100), dtype=float32)
# lab_resh: Tensor("Reshape_1:0", shape=(286, 30), dtype=int32)
osmLoss, checkvalue = osm_loss(final_embedding, rawlabels, centers_embed)
# osmLoss, checkvalue = osm_loss(metric_ftr_in, rawlabels, centers_embed)
SoftMaxloss = model.masked_softmax_cross_entropy(log_resh, lab_resh, msk_resh)
loss = osmLoss
# 为什么loss会固定
# loss = osmLoss
# loss = SoftMaxloss
accuracy = model.masked_accuracy(log_resh, lab_resh, msk_resh)
# optimzie
train_op = model.training(loss, lr, l2_coef)
Path = 'pre_trained/{}/{}/{}'.format(dataset, dataset, self.name)
self.mkdir(Path)
checkpt_file = '{}/allMP_multi_{}_.ckpt'.format(Path, featype)
print('model: {}'.format(checkpt_file))
saver = tf.train.Saver()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
vlss_mn = np.inf
vacc_mx = 0.0
curr_step = 0
with tf.Session(config=config) as sess:
sess.run(init_op)
train_loss_avg = 0
train_acc_avg = 0
val_loss_avg = 0
val_acc_avg = 0
for epoch in range(nb_epochs):
tr_step = 0
tr_size = fea_list[0].shape[0]
# ================ training ============
while tr_step * batch_size < tr_size:
fd1 = {i: d[tr_step * batch_size:(tr_step + 1) * batch_size]
for i, d in zip(ftr_in_list, fea_list)}
fd2 = {i: d[tr_step * batch_size:(tr_step + 1) * batch_size]
for i, d in zip(bias_in_list, biases_list)}
fd3 = {lbl_in: y_train[tr_step * batch_size:(tr_step + 1) * batch_size],
msk_in: train_mask[tr_step * batch_size:(tr_step + 1) * batch_size],
metric_ftr_in: rawFeature,
is_train: True,
attn_drop: 0.6,
ffd_drop: 0.6}
fd = fd1
fd.update(fd2)
fd.update(fd3)
_, loss_value_tr, acc_tr, att_val_train = sess.run([train_op, loss, accuracy, att_val],
feed_dict=fd)
test_check_value = sess.run(checkvalue, feed_dict=fd)
print ("test_check_value: ", test_check_value)
train_loss_avg += loss_value_tr
train_acc_avg += acc_tr
tr_step += 1
vl_step = 0
vl_size = fea_list[0].shape[0]
# ============= val =================
while vl_step * batch_size < vl_size:
# fd1 = {ftr_in: features[vl_step * batch_size:(vl_step + 1) * batch_size]}
fd1 = {i: d[vl_step * batch_size:(vl_step + 1) * batch_size]
for i, d in zip(ftr_in_list, fea_list)}
fd2 = {i: d[vl_step * batch_size:(vl_step + 1) * batch_size]
for i, d in zip(bias_in_list, biases_list)}
fd3 = {lbl_in: y_val[vl_step * batch_size:(vl_step + 1) * batch_size],
msk_in: val_mask[vl_step * batch_size:(vl_step + 1) * batch_size],
metric_ftr_in: rawFeature,
is_train: False,
attn_drop: 0.0,
ffd_drop: 0.0}
fd = fd1
fd.update(fd2)
fd.update(fd3)
loss_value_vl, acc_vl = sess.run([loss, accuracy],
feed_dict=fd)
val_loss_avg += loss_value_vl
val_acc_avg += acc_vl
vl_step += 1
# import pdb; pdb.set_trace()
print('Epoch: {}, att_val: {}'.format(epoch, np.mean(att_val_train, axis=0)))
print('Training: loss = %.5f, acc = %.5f | Val: loss = %.5f, acc = %.5f | vl_step: %d, tr_step: %d' %
(train_loss_avg / tr_step, train_acc_avg / tr_step,
val_loss_avg / vl_step, val_acc_avg / vl_step, vl_step, tr_step))
if val_acc_avg / vl_step >= vacc_mx or val_loss_avg / vl_step <= vlss_mn:
if val_acc_avg / vl_step >= vacc_mx and val_loss_avg / vl_step <= vlss_mn:
vacc_early_model = val_acc_avg / vl_step
vlss_early_model = val_loss_avg / vl_step
saver.save(sess, checkpt_file)
vacc_mx = np.max((val_acc_avg / vl_step, vacc_mx))
vlss_mn = np.min((val_loss_avg / vl_step, vlss_mn))
curr_step = 0
else:
curr_step += 1
if curr_step == patience:
print('Early stop! Min loss: ', vlss_mn,
', Max accuracy: ', vacc_mx)
print('Early stop model validation loss: ',
vlss_early_model, ', accuracy: ', vacc_early_model)
break
train_loss_avg = 0
train_acc_avg = 0
val_loss_avg = 0
val_acc_avg = 0
# check save
saver.save(sess, checkpt_file)
saver.restore(sess, checkpt_file)
print('load model from : {}'.format(checkpt_file))
ts_size = fea_list[0].shape[0]
ts_step = 0
ts_loss = 0.0
ts_acc = 0.0
while ts_step * batch_size < ts_size:
fd1 = {i: d[ts_step * batch_size:(ts_step + 1) * batch_size]
for i, d in zip(ftr_in_list, fea_list)}
fd2 = {i: d[ts_step * batch_size:(ts_step + 1) * batch_size]
for i, d in zip(bias_in_list, biases_list)}
fd3 = {lbl_in: y_all[ts_step * batch_size:(ts_step + 1) * batch_size],
msk_in: all_mask[ts_step * batch_size:(ts_step + 1) * batch_size],
metric_ftr_in: rawFeature,
is_train: False,
attn_drop: 0.0,
ffd_drop: 0.0}
fd = fd1
fd.update(fd2)
fd.update(fd3)
loss_value_ts, acc_ts, jhy_final_embedding, test_final_embeed_check = sess.run([loss, accuracy, final_embedding, test_final_embeed],
feed_dict=fd)
ts_loss += loss_value_ts
ts_acc += acc_ts
ts_step += 1
xx = np.expand_dims(jhy_final_embedding, axis=0)[all_mask]
xx2 = np.expand_dims(test_final_embeed_check, axis=0)[all_mask]
yy = y_all[all_mask]
print ("check fd")
print('xx: {}, yy: {}, ts_size: {}, ts_step: {}, batch_size: {}'.format(xx.shape, yy.shape, ts_size, ts_step,batch_size))
labels, numberofLabels = self.getLabel(yy)
from utils import clustering, pairwise_precision_recall_f1
clusters_pred = clustering(xx2, num_clusters=numberofLabels)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, labels)
print ('prec: ', prec, ', rec: ', rec, ', f1: ', f1, ', originNumberOfClusterlabels: ', numberofLabels)
if needtSNE:
tSNEAnanlyse(xx, labels, join(settings.PIC_DIR, "HAN", "rawReature_%s_final.png" % (self.name)))
tSNEAnanlyse(rawFeature, labels, join(settings.PIC_DIR, "HAN", "rawReature_%s_features.png" % (self.name)))
tSNEAnanlyse(xx2, labels, join(settings.PIC_DIR, "HAN", "rawReature_%s_xx2.png" % (self.name)))
tSNEAnanlyse(xx, clusters_pred, join(settings.PIC_DIR, "HAN", "rawReature_%s_result_label.png" % (self.name)))
sess.close()
return prec, rec, f1, xx2
def gae_for_na(name, rawfeature):
"""
train and evaluate disambiguation results for a specific name
:param name: author name
:return: evaluation results
"""
adj, features, labels = load_local_data(name=name, rawfeature=rawfeature)
# Store original adjacency matrix (without diagonal entries) for later
adj_orig = adj
adj_orig = adj_orig - sp.dia_matrix((adj_orig.diagonal()[np.newaxis, :], [0]), shape=adj_orig.shape)
adj_orig.eliminate_zeros()
adj_train = gen_train_edges(adj)
adj = adj_train
# Some preprocessing
adj_norm = preprocess_graph(adj)
num_nodes = adj.shape[0]
input_feature_dim = features.shape[1]
if FLAGS.is_sparse: # TODO to test
# features = sparse_to_tuple(features.tocoo())
# features_nonzero = features[1].shape[0]
features = features.todense() # TODO
else:
features = normalize_vectors(features)
# Define placeholders
placeholders = {
# 'features': tf.sparse_placeholder(tf.float32),
'features': tf.placeholder(tf.float32, shape=(None, input_feature_dim)),
'adj': tf.sparse_placeholder(tf.float32),
'adj_orig': tf.sparse_placeholder(tf.float32),
'dropout': tf.placeholder_with_default(0., shape=())
}
# Create model
model = None
if model_str == 'gcn_ae':
model = GCNModelAE(placeholders, input_feature_dim)
elif model_str == 'gcn_vae':
model = GCNModelVAE(placeholders, input_feature_dim, num_nodes)
pos_weight = float(adj.shape[0] * adj.shape[0] - adj.sum()) / adj.sum() # negative edges/pos edges
print('positive edge weight', pos_weight)
norm = adj.shape[0] * adj.shape[0] / float((adj.shape[0] * adj.shape[0] - adj.nnz) * 2)
# Optimizer
with tf.name_scope('optimizer'):
if model_str == 'gcn_ae':
opt = OptimizerAE(preds=model.reconstructions,
labels=tf.reshape(tf.sparse_tensor_to_dense(placeholders['adj_orig'],
validate_indices=False), [-1]),
pos_weight=pos_weight,
norm=norm)
elif model_str == 'gcn_vae':
opt = OptimizerVAE(preds=model.reconstructions,
labels=tf.reshape(tf.sparse_tensor_to_dense(placeholders['adj_orig'],
validate_indices=False), [-1]),
model=model, num_nodes=num_nodes,
pos_weight=pos_weight,
norm=norm)
# Initialize session
sess = tf.Session()
sess.run(tf.global_variables_initializer())
adj_label = adj_train + sp.eye(adj_train.shape[0])
adj_label = sparse_to_tuple(adj_label)
def get_embs():
feed_dict.update({placeholders['dropout']: 0})
emb = sess.run(model.z_mean, feed_dict=feed_dict) # z_mean is better
return emb
# Train model
for epoch in range(FLAGS.epochs):
t = time.time()
# Construct feed dictionary
feed_dict = construct_feed_dict(adj_norm, adj_label, features, placeholders)
feed_dict.update({placeholders['dropout']: FLAGS.dropout})
# Run single weight update
outs = sess.run([opt.opt_op, opt.cost, opt.accuracy],
feed_dict=feed_dict)
# Compute average loss
avg_cost = outs[1]
avg_accuracy = outs[2]
print("Epoch:", '%04d' % (epoch + 1), "train_loss=", "{:.5f}".format(avg_cost),
"train_acc=", "{:.5f}".format(avg_accuracy),
"time=", "{:.5f}".format(time.time() - t))
emb = get_embs()
n_clusters = len(set(labels))
emb_norm = normalize_vectors(emb)
clusters_pred = clustering(emb_norm, num_clusters=n_clusters)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, labels)
print('pairwise precision', '{:.5f}'.format(prec),
'recall', '{:.5f}'.format(rec),
'f1', '{:.5f}'.format(f1))
clusters_pred2 = clustering(features, num_clusters=n_clusters)
prec2, rec2, f12 = pairwise_precision_recall_f1(clusters_pred2, labels)
print('pairwise precision', '{:.5f}'.format(prec2),
'recall', '{:.5f}'.format(rec2),
'f1', '{:.5f}'.format(f12))
from sklearn.manifold import TSNE
features_new = TSNE(learning_rate=100).fit_transform(features)
emb_new = TSNE(learning_rate=100).fit_transform(emb_norm)
labels = np.array(labels) + 2
clusters_pred = np.array(clusters_pred) + 2
clusters_pred2 = np.array(clusters_pred2) + 2
if rawfeature == RAW_INTER_NAME:
tSNEAnanlyse(emb_norm, labels, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_final_raw.png" % (name)))
tSNEAnanlyse(features, labels, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_features_raw.png" % (name)))
elif rawfeature == ATTENTIONFEATURE:
tSNEAnanlyse(emb_new, labels, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_final.png" % (name)))
tSNEAnanlyse(features_new, labels, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_features.png" % (name)))
tSNEAnanlyse(emb_new, clusters_pred, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_final_clusterresult.png" % (name)))
tSNEAnanlyse(features_new, clusters_pred2, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_features_clusterresult.png" % (name)))
else:
tSNEAnanlyse(emb_norm, labels, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_final_triplet.png" % (name)))
tSNEAnanlyse(features, labels, join(settings.PIC_DIR, "FINALResult", "rawReature_%s_gae_features_triplet.png" % (name)))
return [prec, rec, f1], num_nodes, n_clusters
def MetricDebug(self, adj_list, fea_list, y_train, y_val, y_test, train_mask, val_mask, test_mask, y_all, all_mask, rawlabels, needtSNE=False, rawFeature=[]):
prec, rec, f1 = 0.0, 0.0, 0.0
nb_nodes = fea_list[0].shape[0]
ft_size = fea_list[0].shape[1]
nb_classes = y_train.shape[1]
# nb_classes = len(set(rawlabels))
# adj = adj.todense()
# features = features[np.newaxis] # [1, nb_node, ft_size]
fea_list = [fea[np.newaxis] for fea in fea_list]
adj_list = [adj[np.newaxis] for adj in adj_list]
y_train = y_train[np.newaxis]
y_val = y_val[np.newaxis]
y_test = y_test[np.newaxis]
y_all = y_all[np.newaxis]
train_mask = train_mask[np.newaxis]
val_mask = val_mask[np.newaxis]
test_mask = test_mask[np.newaxis]
all_mask = all_mask[np.newaxis]
biases_list = [process.adj_to_bias(adj, [nb_nodes], nhood=1) for adj in adj_list]
print('build graph...')
with tf.Graph().as_default():
with tf.name_scope('input'):
metric_ftr_in = tf.placeholder(dtype=tf.float32, shape=(nb_nodes, ft_size), name='metric_ftr_in')
ftr_in_list = [tf.placeholder(dtype=tf.float32,
shape=(batch_size, nb_nodes, ft_size),
name='ftr_in_{}'.format(i))
for i in range(len(fea_list))]
bias_in_list = [tf.placeholder(dtype=tf.float32,
shape=(batch_size, nb_nodes, nb_nodes),
name='bias_in_{}'.format(i))
for i in range(len(biases_list))]
lbl_in = tf.placeholder(dtype=tf.int32, shape=(
batch_size, nb_nodes, nb_classes), name='lbl_in')
msk_in = tf.placeholder(dtype=tf.int32, shape=(batch_size, nb_nodes),
name='msk_in')
attn_drop = tf.placeholder(dtype=tf.float32, shape=(), name='attn_drop')
ffd_drop = tf.placeholder(dtype=tf.float32, shape=(), name='ffd_drop')
is_train = tf.placeholder(dtype=tf.bool, shape=(), name='is_train')
# forward
logits, final_embedding, att_val, centers_embed, test_final_embeed = model.inference(ftr_in_list, nb_classes, nb_nodes, is_train,
attn_drop, ffd_drop,
bias_mat_list=bias_in_list,
hid_units=hid_units, n_heads=n_heads, features=fea_list, labels=rawlabels,
residual=residual, activation=nonlinearity, feature_size=ft_size)
log_resh = tf.reshape(logits, [-1, nb_classes])
lab_resh = tf.reshape(lbl_in, [-1, nb_classes])
msk_resh = tf.reshape(msk_in, [-1])
osm_caa_loss = OSM_CAA_Loss(batch_size=nb_nodes)
osm_loss = osm_caa_loss.forward
osmLoss, checkvalue = osm_loss(final_embedding, rawlabels, centers_embed)
SoftMaxloss = model.masked_softmax_cross_entropy(log_resh, lab_resh, msk_resh)
loss = osmLoss
accuracy = model.masked_accuracy(log_resh, lab_resh, msk_resh)
# optimzie
train_op = model.training(loss, lr, l2_coef)
Path = 'pre_trained/{}/{}/{}'.format(dataset, dataset, self.name)
self.mkdir(Path)
checkpt_file = '{}/allMP_multi_{}_.ckpt'.format(Path, featype)
saver = tf.train.Saver()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
ts_size = fea_list[0].shape[0]
ts_step = 0
ts_loss = 0.0
ts_acc = 0.0
with tf.Session(config=config) as sess:
sess.run(init_op)
saver.restore(sess, checkpt_file)
while ts_step * batch_size < ts_size:
fd1 = {i: d[ts_step * batch_size:(ts_step + 1) * batch_size]
for i, d in zip(ftr_in_list, fea_list)}
fd2 = {i: d[ts_step * batch_size:(ts_step + 1) * batch_size]
for i, d in zip(bias_in_list, biases_list)}
fd3 = {lbl_in: y_all[ts_step * batch_size:(ts_step + 1) * batch_size],
msk_in: all_mask[ts_step * batch_size:(ts_step + 1) * batch_size],
metric_ftr_in: rawFeature,
is_train: False,
attn_drop: 0.0,
ffd_drop: 0.0}
fd = fd1
fd.update(fd2)
fd.update(fd3)
test_final_embeed_check = sess.run([ test_final_embeed], feed_dict=fd)
ts_step += 1
xx2 = np.expand_dims(test_final_embeed_check, axis=0)[all_mask]
yy = y_all[all_mask]
labels, numberofLabels = self.getLabel(yy)
from utils import clustering, pairwise_precision_recall_f1
clusters_pred = clustering(xx2, num_clusters=numberofLabels)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, labels)
print ('prec: ', prec, ', rec: ', rec, ', f1: ', f1, ', originNumberOfClusterlabels: ', numberofLabels)
if needtSNE:
tSNEAnanlyse(rawFeature, labels, join(settings.PIC_DIR, "MetricLearning", "rawReature_%s_features.png" % (self.name)))
tSNEAnanlyse(xx2, labels, join(settings.PIC_DIR, "MetricLearning", "rawReature_%s_xx2.png" % (self.name)))
def train(name, needtSNE=False, savefile=True):
adj, adj2, features, labels, Clusterlabels, Ids = load_local_data(
name=name)
initClusterlabel = Clusterlabels
oneHotClusterLabels = toOneHot(Clusterlabels)
num_logits = len(oneHotClusterLabels[0])
# enc.transform([['Female', 1], ['Male', 4]]).toarray()
print('debuging ', oneHotClusterLabels.shape)
originClusterlabels = Clusterlabels
n_clusters = len(set(labels))
OldClusterlabels = Clusterlabels
originNumberOfClusterlabels = len(set(Clusterlabels))
num_nodes = adj.shape[0]
input_feature_dim = features.shape[1]
adj_norm, adj_label = NormalizedAdj(adj)
adj_norm2, adj_label2 = NormalizedAdj(adj2)
if FLAGS.is_sparse: # TODO to test
# features = sparse_to_tuple(features.tocoo())
# features_nonzero = features[1].shape[0]
features = features.todense() # TODO
else:
features = normalize_vectors(features)
# Define placeholders
placeholders = {
# 'features': tf.sparse_placeholder(tf.float32),
'features': tf.placeholder(tf.float32,
shape=(None, input_feature_dim)),
'labels': tf.placeholder(tf.int64, shape=(None), name='labels'),
'graph1': tf.sparse_placeholder(tf.float32),
'graph2': tf.sparse_placeholder(tf.float32),
'graph1_orig': tf.sparse_placeholder(tf.float32),
'graph2_orig': tf.sparse_placeholder(tf.float32),
'dropout': tf.placeholder_with_default(0., shape=()),
'epoch': tf.placeholder_with_default(0., shape=()),
'clusterEpoch': tf.placeholder_with_default(0., shape=())
}
# pos_weight = float(adj.shape[0] * adj.shape[0] - adj.sum()) / adj.sum() # negative edges/pos edges
# norm = adj.shape[0] * adj.shape[0] / float((adj.shape[0] * adj.shape[0] - adj.nnz) * 2)
def get_embs():
feed_dict.update({placeholders['dropout']: 0})
emb = sess.run(model.z_mean_1, feed_dict=feed_dict) # z_mean is better
return emb
def getGraphDetail(adj):
pos_weight = float(adj.shape[0] * adj.shape[0] -
adj.sum()) / adj.sum() # negative edges/pos edges
norm = adj.shape[0] * adj.shape[0] / float(
(adj.shape[0] * adj.shape[0] - adj.nnz) * 2)
return {'norm': norm, 'pos_weight': pos_weight}
# return pos_weight, norm
# loss1s = []
# loss2s = []
# loss3s = []
n_clusters = len(set(labels))
graph1 = getGraphDetail(adj)
graph2 = getGraphDetail(adj2)
# construct adj_orig
graph1['labels'] = tf.reshape(
tf.sparse_tensor_to_dense(placeholders['graph1_orig'],
validate_indices=False), [-1])
graph2['labels'] = tf.reshape(
tf.sparse_tensor_to_dense(placeholders['graph2_orig'],
validate_indices=False), [-1])
# Train model
for clusterepoch in range(FLAGS.clusterEpochs):
print('cluster epoch: ', clusterepoch)
# tf.reset_default_graph()
# num_logits
model = BuildModel(placeholders,
input_feature_dim,
num_nodes,
name='model%d' % (clusterepoch),
num_logits=num_logits)
# Session
# tf.reset_default_graph()
# sess = tf.InteractiveSession()
opt = OptimizerDualGCNAutoEncoder(model=model,
num_nodes=num_nodes,
z_label=Clusterlabels,
name='model%d' % (clusterepoch),
graph1=graph1,
graph2=graph2)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
# Centers
# centers = opt.centers
for epoch in range(FLAGS.epochs):
# print ('epoch: ', epoch)
# opt.epoch = epoch
model.epoch = epoch
# Construct feed dictionary
# Number of logics and preb
feed_dict = construct_feed_dict(adj_norm, adj_label, adj_norm2,
adj_label2, features, placeholders,
Clusterlabels, epoch,
clusterepoch + 1)
feed_dict.update({placeholders['dropout']: FLAGS.dropout})
# Run single weight update
# outs = sess.run([opt.opt_op, opt.cost, opt.accuracy], feed_dict=feed_dict)
outs = sess.run([opt.opt_op, opt.cost], feed_dict=feed_dict)
# [Loss, softmax_loss, loss3, centerloss, reconstructloss] = sess.run([opt.cost, opt.softmax_loss, opt.loss3, opt.centerloss, opt.reconstructloss], feed_dict=feed_dict)
# [Loss, loss3, centerloss, reconstructloss, L2loss] = sess.run([opt.cost, opt.loss3, opt.centerloss, opt.reconstructloss, opt.L2loss], feed_dict=feed_dict)
[Loss, reconstructloss] = sess.run([opt.cost, opt.reconstructloss],
feed_dict=feed_dict)
# print ('loss: ', Loss, ', loss1: ', loss1, ', loss2: ', loss2 ,', centerloss: ', centerloss, ', acc: ', outs[2])
print('epoch: ', epoch, '， loss: ', Loss, ', reconstructloss : ',
reconstructloss)
# if clusterepoch != FLAGS.clusterEpochs -1 :
emb = get_embs()
X_new = TSNE(learning_rate=100).fit_transform(emb)
tClusterLabels = []
Maxscore = -10000
NumberOfCluster = 0
for nc in range(2, originNumberOfClusterlabels + 1, 1):
TempLabels = clustering(X_new, nc)
score = silhouette_score(X_new, TempLabels)
print('nc: ', nc, ', score: ', score)
if score > Maxscore:
Maxscore = score
tClusterLabels = TempLabels
NumberOfCluster = nc
print('NumberOfCluster: ', NumberOfCluster,
', originNumberOfClusterlabels : ', originNumberOfClusterlabels,
', Maxscore: ', Maxscore)
if NumberOfCluster < 0 or NumberOfCluster > originNumberOfClusterlabels:
continue
# 符合不断缩小的要求
# 重新修改这些参数
Clusterlabels = tClusterLabels
originNumberOfClusterlabels = NumberOfCluster
prec, rec, f1 = pairwise_precision_recall_f1(Clusterlabels, labels)
print('prec: ', prec, ', rec: ', rec, ', f1: ', f1,
', originNumberOfClusterlabels: ', originNumberOfClusterlabels)
Cc = Counter(Clusterlabels)
print(Cc)
if needtSNE:
sNEComparingAnanlyse(emb,
OldClusterlabels,
labels,
Clusterlabels,
savepath=join(
settings.PIC_DIR,
"%s_%s.png" % (name, clusterepoch)))
# tSNEAnanlyse(emb, labels, join(settings.PIC_DIR, "%s.png"%(clusterepoch)) )
# tf.reset_default_graph()
emb = get_embs()
emb_norm = normalize_vectors(emb)
clusters_pred = clustering(emb_norm,
num_clusters=originNumberOfClusterlabels)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, labels)
print('prec: ', prec, ', rec: ', rec, ', f1: ', f1,
', originNumberOfClusterlabels: ', originNumberOfClusterlabels)
# lossPrint(range(FLAGS.epochs), loss1s, loss2s, loss3s)
if needtSNE:
tSNEAnanlyse(emb, labels,
join(settings.PIC_DIR, "%s_final.png" % (name)))
tf.reset_default_graph()
return [prec, rec, f1], num_nodes, n_clusters
while epoch < epochs:
_, losscheck, value2 = sess.run([train_op, loss, checkvalue], feed_dict=fd)
print ("epoch: {} loss: {}, checkvalue: {}".format(epoch, losscheck, value2))
epoch += 1
print ("final_embed: ", final_embed)
embedding = sess.run([final_embed], feed_dict=fd)
embedding = embedding[0]
print ("embedding: ", embedding)
from utils import clustering, pairwise_precision_recall_f1
clusters_pred = clustering(embedding, num_clusters=nb_class)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, rawlabels)
print ('prec: ', prec, ', rec: ', rec, ', f1: ', f1, ', originNumberOfClusterlabels: ', nb_class)
tSNEAnanlyse(embedding, rawlabels, join(settings.PIC_DIR, "PureMetricLoss", "%s_final.png" % (name)))
tSNEAnanlyse(features, rawlabels, join(settings.PIC_DIR, "PureMetricLoss", "%s_features.png" % (name)))
# my_KNN(xx, yy)
# my_Kmeans(xx, yy)
sess.close()
labels.append(aid)
rf.append(rawFeature.get(pid))
tf.append(tripletFeature.get(pid))
attentionf.append(lc_emb.get(pid))
labels = encode_labels(labels)
numberofLabels = len(set(labels))
def clusterTest(embedding, numberofLabels):
clusters_pred = clustering(embedding, num_clusters=numberofLabels)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, labels)
return [prec, rec, f1]
tSNEAnanlyse(rf, labels, join(settings.PIC_DIR, "FINALResult", "%s_rawFeature.png" % (name)))
tSNEAnanlyse(tf, labels, join(settings.PIC_DIR, "FINALResult", "%s_tripletFeature.png" % (name)))
tSNEAnanlyse(attentionf, labels, join(settings.PIC_DIR, "FINALResult", "%s_lcmbFeature.png" % (name)))
Res = {}
Res['rawfeature'] = clusterTest(rf, numberofLabels=numberofLabels)
Res['tripletfeature'] = clusterTest(tf, numberofLabels=numberofLabels)
Res['lcmbfeature'] = clusterTest(attentionf, numberofLabels=numberofLabels)
print ("Res: ", Res)
EndIndex = -2
featurePath = getPATH(name, idf_threshold, 'feature_and_label', ispretrain)
# idx_features_labels = np.genfromtxt(join(settings.DATA_DIR, 'local', 'graph-{}'.format(idf_threshold)), dtype=np.dtype(str))
idx_features_labels = np.genfromtxt(featurePath, dtype=np.dtype(str))
features = np.array(idx_features_labels[:, 1:EndIndex],
dtype=np.float32) # sparse?
rawlabels = encode_labels(idx_features_labels[:, EndIndex])
pids = idx_features_labels[:, 0]
return features, pids, rawlabels
def load_test_names():
return data_utils.load_json(settings.DATA_DIR, 'test_name_list2.json')
Res = {}
names = load_test_names()
for name in names:
features, pids, rawlabels = loadFeature(name, ispretrain=False)
tSNEAnanlyse(
features, rawlabels,
join(settings.PIC_DIR, "MetricLearning",
"rawReature_%s_train.png" % (name)))
numberofLabels = len(set(rawlabels))
clusters_pred = clustering(features, num_clusters=numberofLabels)
prec, rec, f1 = pairwise_precision_recall_f1(clusters_pred, rawlabels)
Res[name] = {"prec": prec, "rec": rec, "f1": f1}
print(Res)
embs_input = []
labels = []
pids = []
for i, aid in enumerate(name_data.keys()):
if len(name_data[aid]) < 5: # n_pubs of current author is too small
continue
for pid in name_data[aid]:
cur_emb = lc_input.get(pid)
if cur_emb is None:
continue
embs_input.append(cur_emb)
pids.append(pid)
labels.append(aid)
embs_input = np.stack(embs_input)
inter_embs = get_hidden_output(trained_global_model, embs_input)
labels = encode_labels(labels)
for i, pid_ in enumerate(pids):
res_embs.append(inter_embs[i])
# Clustering and save the result
tSNEAnanlyse(
res_embs, labels,
join(settings.PIC_DIR, "OnlyTriplete",
"rawReature_%s_triplet.png" % (name)))
tSNEAnanlyse(
embs_input, labels,
join(settings.PIC_DIR, "OnlyTriplete",
"rawReature_%s_features.png" % (name)))